U.S. patent application number 17/084327 was filed with the patent office on 2022-05-05 for data-driven sales recommendation tool.
The applicant listed for this patent is EMC IP Holding Company LLC. Invention is credited to Avitan Gefen, Noga Gershon, Amihai Savir.
Application Number | 20220138820 17/084327 |
Document ID | / |
Family ID | |
Filed Date | 2022-05-05 |
United States Patent
Application |
20220138820 |
Kind Code |
A1 |
Gershon; Noga ; et
al. |
May 5, 2022 |
DATA-DRIVEN SALES RECOMMENDATION TOOL
Abstract
One example method includes receiving a quote for provision of
goods and/or services, and the quote concerns a particular account,
receiving information concerning characteristics of the account
identified in the quote, receiving information concerning
characteristics of the goods and/or services specified in the
quote, generating a probability that the quote will be approved by
the account, and the probability is generated based on the
characteristics of the account and the characteristics of the goods
and/or services specified in the quote, when the probability is
below a threshold, generating an adjusted quote based on input
received, and generating an updated probability based on the
adjusted quote.
Inventors: |
Gershon; Noga; (Dimona,
IL) ; Savir; Amihai; (Sansana, IL) ; Gefen;
Avitan; (Tel Aviv, IL) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
EMC IP Holding Company LLC |
Hopkinton |
MA |
US |
|
|
Appl. No.: |
17/084327 |
Filed: |
October 29, 2020 |
International
Class: |
G06Q 30/06 20060101
G06Q030/06; G06N 20/00 20060101 G06N020/00; G06Q 10/06 20060101
G06Q010/06; G06F 17/18 20060101 G06F017/18 |
Claims
1. A method, comprising: receiving a quote for provision of goods
and/or services, and the quote concerns a particular account;
receiving information concerning characteristics of the account
identified in the quote; receiving information concerning
characteristics of the goods and/or services specified in the
quote; generating a probability that the quote will be approved by
the account, and the probability is generated based on the
characteristics of the account and the characteristics of the goods
and/or services specified in the quote; when the probability is
below a threshold, generating an adjusted quote based on input
received; and generating an updated probability based on the
adjusted quote.
2. The method as recited in claim 1, wherein part of the method
performed by a machine learning model.
3. The method as recited in claim 1, wherein the probability and/or
updated probability are generated based on one or both of: a first
deal cluster comprising a plurality of deals and identifying common
quote characteristics among respective quotes that are associated
with the deals; and/or a second deal cluster comprising a plurality
of deals and identifying common account characteristics among
respective accounts associated with the deals.
4. The method as recited in claim 3, further comprising assigning
the updated quote to one of the deal clusters, and updating a ratio
of approved deals associated with the deal cluster to which the
updated quote is assigned.
5. The method as recited in claim 1, further comprising:
presenting, to a user, possible modifications to goods and/or
services specified in the quote; receiving, from the user, the
input, and the input indicating selection of one or more of the
modifications; generating the adjusted quote based on the input
received from the user; and providing, to the user, the updated
probability.
6. The method as recited in claim 5, wherein the modifications are
presented in serial form, or hierarchical form.
7. The method as recited in claim 1, wherein the updated
probability and/or the updated quote are generated almost
immediately after receipt of the input from the user.
8. The method as recited in claim 1, further comprising: generating
a first deal cluster comprising a plurality of deals and
identifying common quote characteristics among respective quotes
that are associated with the deals; and/or generating a second deal
cluster comprising a plurality of deals and identifying common
account characteristics among respective accounts associated with
the deals, and generation of the updated probability is based on
the characteristics identified in the first deal cluster and/or the
characteristics identified in the second deal cluster.
9. The method as recited in claim 1, wherein a margin associated
with the quote is maintained in the updated quote.
10. The method as recited in claim 1, wherein the updated
probability is higher than the probability.
11. A non-transitory storage medium having stored therein
instructions that are executable by one or more hardware processors
to perform operations comprising: receiving a quote for provision
of goods and/or services, and the quote concerns a particular
account; receiving information concerning characteristics of the
account identified in the quote; receiving information concerning
characteristics of the goods and/or services specified in the
quote; generating a probability that the quote will be approved by
the account, and the probability is generated based on the
characteristics of the account and the characteristics of the goods
and/or services specified in the quote; when the probability is
below a threshold, generating an adjusted quote based on input
received; and generating an updated probability based on the
adjusted quote.
12. The non-transitory storage medium as recited in claim 11,
wherein one or more of the operations are performed by a machine
learning model.
13. The non-transitory storage medium as recited in claim 11,
wherein the probability and/or updated probability are generated
based on one or both of: a first deal cluster comprising a
plurality of deals and identifying common quote characteristics
among respective quotes that are associated with the deals; and/or
a second deal cluster comprising a plurality of deals and
identifying common account characteristics among respective
accounts associated with the deals.
14. The non-transitory storage medium as recited in claim 13,
wherein the operations further comprise assigning the updated quote
to one of the deal clusters, and updating a ratio of approved deals
associated with the deal cluster to which the updated quote is
assigned.
15. The non-transitory storage medium as recited in claim 11,
wherein the operations further comprise: presenting, to a user,
possible modifications to goods and/or services specified in the
quote; receiving, from the user, the input, and the input
indicating selection of one or more of the modifications;
generating the adjusted quote based on the input received from the
user; and providing, to the user, the updated probability.
16. The non-transitory storage medium as recited in claim 15,
wherein the modifications are presented in serial form, or
hierarchical form.
17. The non-transitory storage medium as recited in claim 11,
wherein the updated probability and/or the updated quote are
generated almost immediately after receipt of the input from the
user.
18. The non-transitory storage medium as recited in claim 11,
wherein the operations further comprise: generating a first deal
cluster comprising a plurality of deals and identifying common
quote characteristics among respective quotes that are associated
with the deals; and/or generating a second deal cluster comprising
a plurality of deals and identifying common account characteristics
among respective accounts associated with the deals, and generation
of the updated probability is based on the characteristics
identified in the first deal cluster and/or the characteristics
identified in the second deal cluster.
19. The non-transitory storage medium as recited in claim 11,
wherein a margin associated with the quote is maintained in the
updated quote.
20. The non-transitory storage medium as recited in claim 11,
wherein the updated probability is higher than the probability.
Description
FIELD OF THE INVENTION
[0001] Embodiments of the present invention generally relate to
data analysis. More particularly, at least some embodiments of the
invention relate to systems, hardware, software, computer-readable
media, and methods for using both historical data and
user-specified constraints to generate a data set having a
corresponding probability of being accepted by the user.
BACKGROUND
[0002] In modern competitive markets, whether for services or
products, it can be important for an enterprise to be able to
anticipate the needs of its customers precisely and quickly. A
mistaken understanding of customer needs, and/or a failure to
timely understand those needs, could result in lost opportunities
for the enterprise. To illustrate, a salesman may have a limited
ability to evaluate the likelihood that a particular proposal would
be approved by a customer. This limited ability may be due, for
example, to a lack of evaluative capabilities and resources, and/or
to a lack of relevant data. Further, even where a salesman may be
able to formulate a proposal, modification of the proposal, with
the aim of the proposal being approved according to some
constrains, can be a complex task which requires time as well as a
deep understanding of the various different components of the
proposal.
BRIEF DESCRIPTION OF THE DRAWINGS
[0003] In order to describe the manner in which at least some of
the advantages and features of the invention may be obtained, a
more particular description of embodiments of the invention will be
rendered by reference to specific embodiments thereof which are
illustrated in the appended drawings. Understanding that these
drawings depict only typical embodiments of the invention and are
not therefore to be considered to be limiting of its scope,
embodiments of the invention will be described and explained with
additional specificity and detail through the use of the
accompanying drawings.
[0004] FIG. 1 discloses aspects of a machine learning (ML) model
according to some example embodiments.
[0005] FIG. 2 discloses aspects of an example scenario involving
determination of a probability of acceptance.
[0006] FIG. 3 discloses aspects of another example scenario
involving determination of a probability of acceptance.
[0007] FIG. 4 is a flow diagram disclosing aspects of an example
method.
[0008] FIG. 5 discloses aspects of an example computing entity that
may be configured to perform any of the disclosed methods and
processes.
DETAILED DESCRIPTION OF SOME EXAMPLE EMBODIMENTS
[0009] Embodiments of the present invention generally relate to
data analysis. More particularly, at least some embodiments of the
invention relate to systems, hardware, software, computer-readable
media, and methods for using both historical data and
user-specified constraints to generate a data set having a
corresponding probability of being accepted by the user. Some
particular embodiments are directed to prediction of the
probability that a particular proposed agreement will be accepted
by a party, although the scope of the invention is not limited to
this particular application.
[0010] In general, at least some example embodiments of the
invention may employ a data driven approach that applies a machine
learning model on historical data and predicts the probability that
a proposed agreement, or quote, generated by a sales entity will be
accepted by a particular customer entity. The sales entity may be
associated with an enterprise that offers products and/or services
for sale to one or more customers. In some instances, a customer
entity may comprise a computing system programmed, for example, to
evaluate the quote in light of such parameters as
customer-specified constraints.
[0011] In cases where the probability of the acceptance of a quote
proposal may be relatively low, the system may recommend actions,
such as modification of the quote proposal, to be taken in order to
increase the attractiveness of the offer so as to correspondingly
increase the probability that the quote will be accepted by the
customer entity. Such actions may include, for example, upgrading
some products, and/or replacing items with equivalent ones so that
a bigger discount can be given. These actions may be selected on
various bases, such as with a view to minimizing the impact on an
expected margin associated with the quote, and/or with regards to
constraints imposed by the sales entity and/or the customer entity.
Example embodiments may employ various mechanisms to these ends. To
illustrate, example embodiments may employ a machine learning
model, which may be based on and/or employ historical data, that
may help to estimate the probability that a particular quote will
be approved by the customer entity to whom the quote is proposed.
Another mechanism that may be employed in some embodiments is a
mechanism that may help construct a better quote based on
constraints imposed by the customer, as well as constraints imposed
by the sales entity. A proposal may embrace a proposed quote or
transaction between/among two or more parties, such as a sales
entity and customer entity, for example. A proposal may, or may
not, actually be implemented between/among the parties.
[0012] Embodiments of the invention, such as the examples disclosed
herein, may be beneficial in a variety of respects. For example,
and as will be apparent from the present disclosure, one or more
embodiments of the invention may provide one or more advantageous
and unexpected effects, in any combination, some examples of which
are set forth below. It should be noted that such effects are
neither intended, nor should be construed, to limit the scope of
the claimed invention in any way. It should further be noted that
nothing herein should be construed as constituting an essential or
indispensable element of any invention or embodiment. Rather,
various aspects of the disclosed embodiments may be combined in a
variety of ways so as to define yet further embodiments. Such
further embodiments are considered as being within the scope of
this disclosure. As well, none of the embodiments embraced within
the scope of this disclosure should be construed as resolving, or
being limited to the resolution of, any particular problem(s). Nor
should any such embodiments be construed to implement, or be
limited to implementation of, any particular technical effect(s) or
solution(s). Finally, it is not required that any embodiment
implement any of the advantageous and unexpected effects disclosed
herein.
[0013] In particular, one advantageous aspect of at least some
embodiments of the invention is that an embodiment, such as a sales
entity, may employ historical information concerning prior dealings
with a customer entity and/or constraints imposed by the customer
entity to generate a proposal with a relatively high, or relatively
higher, likelihood of acceptance by the customer entity. An
embodiment may measure the likelihood of acceptance with reference
to a scale such as a numerical scale, and/or with respect to the
respective likelihoods of acceptance of one or more other proposals
that may have been generated by the sales entity, by the customer
entity. An embodiment may enable modification of an existing
proposal to create a proposal with a relatively higher likelihood
of acceptance. An embodiment of the invention may take into
consideration, such as in the form of an input to a computer
performed process, human insights and/or other human-generated data
as a basis for generating a new or modified proposal. Embodiments
of the invention embrace methods and processes that are beyond the
ability of a human to perform, practically or even at all, as a
mental process, and such methods and processed performed in
connection with embodiments of the invention may employ various
combinations of factors and data to generate new/modified
proposals. As such, these methods and processes may be performed
quickly, effectively, and without introduction of human error in
such performance.
A. Overview
[0014] The following is a discussion of aspects of example
operating environments for various embodiments of the invention.
This discussion is not intended to limit the scope of the
invention, or the applicability of the embodiments, in any way.
[0015] Competition today is as fierce as ever due to the abundance
of competitors in the goods and services markets. In order to stay
a major contestant in this competitive landscape, companies should
offer their customers the best possible prices, quantity and
quality of goods and services before their competitors are able to
do so. Typically, however, salespersons use only their own human
intuition to estimate the likelihood that a proposal will be
accepted by a customer. Offering a proposal that may be relatively
likely to be rejected might waste precious time and lead to the
acceptance, by the target customer, of a better deal made by a
competitor company.
[0016] In more detail, a salesman typically relies primarily, or
even exclusively, on his intuition in constructing a quote that,
the salesman hopes, will approved by the customer. This approach
may be rather tricky however, since intuition is different between
people and may grow, develop, and change, as experience is gained
by the salesman. An incorrect assessment of the likelihood that a
quote will be approved by the prospective customer may result in
lost opportunities. For example, while one offer is being rejected,
a more appealing one, such as from a competitor, might appear and
get approval from the customer before the initial offer could be
adjusted to better meet the needs of the customer.
[0017] As well, it is a complex task to adjust parameters of a
particular proposal, while still maintaining, for example, a
desired margin, that is, a profit margin, for the proposal. In more
detail, many proposals that are offered to customers are complex
and composed of a variety of different components. Maintenance of a
margin in an updated quote embraces any margin within plus/minus
about 3% of the margin associated with the initial quote.
[0018] In addition, there may be many ways to improve a certain
proposal without unduly reducing the margin associated with the
proposal. However, the salesman may not necessarily have access to,
or even be aware of the existence of, information concerning those
possible avenues to improvement. Thus, the salesman may have little
recourse in terms of the data that is available to be used as a
basis for adjusting a proposal. In some cases, for example, the
desired margin for the proposal may be the only information
available to the salesman. A simplistic approach to proposal
construction/modification, such as one that relies largely or
exclusively on margin information, will likely be inadequate to
produce a proposal with an acceptable likelihood of acceptancy by
the target customer.
[0019] Moreover, manual adjustment to a proposal, such as may be
performed by a salesman, will likely not produce acceptable results
since a manual process is too slow, complex, and/or susceptible to
human error, to generate the proposal in an accurate and timely
manner. For example, a manual process would require a trial and
error, rather than systematic, approach, to proposal
creation/modification. Such a trial and error approach may prolong
the time between the moment the target customer approached the
company of the salesman and the moment an offer is proposed to the
target customer. As noted earlier, such a delay may lead to lost
sales opportunities.
B. Aspects of Some Example Embodiments
[0020] With the foregoing considerations in mind, attention is
directed now to aspects of some example embodiments. As noted
earlier herein, embodiments of the invention may facilitate a
process of constructing a quote, or proposal, while maximizing, or
at least improving, the likelihood of approval of the proposal by
the customer. Some example embodiments may include at least two
mechanisms, namely, a machine learning model based on historical
data that may help to estimate the probability that a particular
proposal will be approved by the customer, and a second mechanism
that may help construct a better quote, that is, one with a
relatively high/higher likelihood of approval by a target customer,
based on constraints imposed by the customer, and/or constraints
imposed by the salesman. Thus, the second mechanism may be a
mechanism that uses constraints and/or other input provided by the
customer entity and/or the sales entity.
[0021] Embodiments of the invention may facilitate the process of
constructing a proposal, while also maximizing, or at least
improving, the odds of the approval of the proposal by the
customer. Doing so can increase the competitiveness of the
enterprise, and may increase the trust and satisfaction of the
customers of the customer.
[0022] As described above, embodiments may employ two mechanisms
that may support a process for generating/modifying a proposal. In
some instances, each mechanism may employ a different respective
data set. For the mechanism of predicting the probability that a
proposal will be accepted, quote data may be employed which may
contain information about product characteristics desired by the
customer, as well as data concerning the account characteristics,
that is, characteristics of the target customer for the proposal.
The second mechanism may use data such as engineered product
features of products produced by the enterprise. Because the extent
to which product characteristics desired by the customer match
those product features actually available from the enterprise may
vary, the second mechanism can attempt to find the best match
between what the customer wants and what is actually available.
This concept may be illustrated by various products provided by
Dell
(https://www.dell.com/en-us/shop/dell-laptops/inspiron-14-3000-laptop/spd-
/inspiron-14-3480-laptop), where each available product and its
selectable features are shown. Because the features may be
individually selectable, a laptop, for example, may be defined that
has a customized configuration specified by the user or
purchaser.
[0023] It was noted earlier that various data may be used in
connection with embodiments of the invention. As used herein, the
term `data` is intended to be broad in scope. Thus, that term
embraces, by way of example and not limitation, data segments, data
chunks, data blocks, atomic data, emails, objects of any type,
files of any type including media files such as audio files and
video files, word processing files, spreadsheet files, and database
files, as well as contacts, directories, sub-directories, volumes,
and any group of one or more of the foregoing. Example embodiments
of the invention are applicable to any system capable of storing
and handling various types of objects, in analog, digital, or other
form. Although terms such as document, file, segment, block, or
object may be used by way of example, the principles of the
disclosure are not limited to any particular form of representing
and storing data or other information. Rather, such principles are
equally applicable to any object capable of representing
information.
[0024] Finally, as used herein, a `quote` or `proposal` refers to a
potential transaction that may be proposed by a sales entity to a
customer entity. Further, a `deal` refers to a transaction that has
been approved, that is, agreed to, by a customer. Thus, a quote or
proposal may, or may not, ultimately end in a deal.
[0025] With particular attention now to FIG. 1, a scheme 100 is
disclosed that may serve to generate new/modified proposals. In
general, the example scheme 100 may include a machine learning
model 102 that is operable to receive various inputs 103 based on
which the machine learning model 102 may generate a probability
score 104 that reflects an extent to which it is probable that a
particular proposal will be accepted by a customer target. In the
example of FIG. 1, a relatively high probability 106 may suggest
that the proposal be submitted to the target customer, while a
relatively low probability 108 may suggest that one or more aspects
of the proposal should be adjusted to attempt to improve the
probability, that is, an updated proposal 110 may be generated if
the probability of acceptance is determined to be low or
unacceptable.
[0026] The probability may, but need not necessarily be, expressed
as a numerical value, such as a number in a range of 0 to 10, where
zero indicates no probability that a proposal will be accepted, and
10 indicates certainty that the proposal will be accepted. A `high`
or `acceptable` probability may be defined, for example, as a
probability with a value of 7 or higher, and a `low` or
`unacceptable` probably may be defined, for example, as a
probability with a value of 6 or lower. The range, and expression,
of probabilities may be embodied in any other suitable form, and
the foregoing is provided only by way of example and is not
intended to limit the scope of the invention in any way.
[0027] B.1 Aspects of an Example Machine Learning Model
[0028] With continued reference to FIG. 1, various data and
processes 200 may be used to generate, and/or otherwise obtain and
provide, input to the machine learning model 102. Such data may
include historical data 202 which may be subjected to data
preprocessing 204, and the historical data 202 may be used as a
basis to engineer features 206 that may be included in products
offered to customers. For example, if the historical data suggests
that customers prefer hard drives of at least 1 Tb, the enterprise
may decide to manufacture products that include hard drives of at
least that size. The engineered features 206 may also be created
and/or implemented based on order and/or account characteristics
208. For example, if the customer handles sensitive data, the
customer may need data security features in any products that it
purchases. The engineered features 206, which may be features
available in the products of the enterprise, along with historical
data 202, and/or, order/account characteristics, may all serve as
inputs to the machine learning model 102.
[0029] With continued reference to FIG. 1, further details are
provided concerning the example machine learning model 102. In
order to build the machine learning model 102 for predicting the
probability of the acceptance of a proposal, the quote
characteristics identifying the needs of the customer,
client/account characteristics 208, as well as similarity-based
engineered features 206, and historical data 202, may be used as
inputs to the model. The historical data 202 may be used as a
direct input to the machine learning model 102 and/or as an input
that may be used to inform development and engineering of the
similarity-based engineered features 206.
[0030] As noted earlier, various historical data 202 may be
employed as an input to a machine learning model 102. In at least
some embodiments, historical data may comprise characteristics of
one or more prior quotes, or proposals, that are associated with
respective deals, and the historical data may additionally, or
alternatively, comprise characteristics of one or more clients or
customers that are associated with respective deals.
[0031] In more detail, deals may be clustered together in various
ways. For example, body of input data for the machine learning
model 102 may comprise one or more clusters of deals, where the
clusters are defined based on similarities among quote
characteristics of the quotes that underly those deals. That is,
each deal in a group of deals may be associated with a respective
set of one more quotes. Where quotes have similar characteristics,
those quotes and, thus, the deals with which those quotes are
associated, may be clustered together.
[0032] As another example, deals may be clustered together based on
similarities among the accounts with which those deals are
associated. That is, deals whose respective associated accounts
have a particular degree of similarity with each other, may be
clustered together.
[0033] The scope of the invention is not limited to the use of deal
clusters based on quote characteristics, and account
characteristics. More generally, deals may be clustered on any
other additional, or alternative, bases, and the foregoing are
provided only by way of example.
[0034] Either or both of the clusters concerning quote
characteristics, and account characteristics, respectively, may be
used to train the machine learning model 102 so that when a new
quote or proposal is provided to the machine learning model 102,
the machine learning model 102 may resort to these clusters of
training data when generating a probability of acceptance for the
new quote.
[0035] With reference first to quotes, clustering of deals based on
quote information may be used to identify and capture similarity
between quotes. For example, quotes may be determined to be similar
to each other based on the products or services to which they
relate, so that, for example, quotes relating to the same product,
or service, may be deemed to be similar to each other.
[0036] For each deal in a cluster, to which one, some, or all, of
the quotes relate, a ratio may be calculated of (deals)/(quotes).
Such a deal ratio, or ratio of deals, may indicate, for example,
which deals are reached relatively quickly in terms of the number
of quotes needed before a deal was reached, and which deals took
relatively longer to reach. In general, a ratio of 1 may be ideal,
as it would indicate only 1 quote was needed to reach the deal,
while a ratio of 0.2 for example, indicates that 5 quotes were
needed before the deal was reached. Thus, the machine learning
model 102 may choose to start with, or at least consider, the quote
that corresponds to the ratio of 1, rather than the quote that
corresponds to the ratio of 0.2, since the former quote was
accepted more quickly. Because deals may be clustered together
according to the similarity of the respective quote(s) to which
they correspond, a comparison of the respective associated deal
ratios for each deal may be useful.
[0037] When a new quote is generated, the new quote may be assigned
to one of the deal clusters, based on similarity between the new
quote and the quotes of the cluster. As well, an overall deal ratio
of that cluster (total deals of cluster)/(total quotes in cluster)
may be assigned to the new quote.
[0038] In addition to quote characteristics, account
characteristics and similarities may also be used as an input to
the machine learning model 102. That is, account-based training
data for the machine learning model 102 may be generated by
clustering deals together based on similarities among the
characteristics of the respective accounts associated with those
deals.
[0039] Thus, the training data may comprise the characteristics of
various accounts, and the extent to which accounts may be similar
to each other. Similarity of accounts may be determined on any
suitable basis. For example, accounts may be determined to be
similar if they each have a need for a particular product or
service, if they each handle a particular type of data, or on any
additional or alternative bases. More generally, similarity between
accounts, and between quotes, may be defined based on any
characteristic(s).
[0040] After the deals have been clustered based on the similarity
of their respective underlying accounts, similarity measures among
the deals in each cluster may be calculated. Calculation of
similarity measures may be performed in any suitable way, and on
any suitable bases. For example, if the characteristics of an
account associated with a first deal are the same as the
characteristics of an account associated with a second deal, which
may be the case when the two accounts are actually the same
account, then the similarity between the first and second deals may
be 1.0. On the other hand, if the two accounts have 3
characteristics, out of a total of 7 each, that are the same, the
similarity between the two deals may be expressed as ( 3/7) or 0.4.
The similarity may be refined by weighting the characteristics of
the accounts such that, for example, the fact that two accounts
operate in the same markets may have relatively less weight than
the fact that the two accounts typically buy the same types and
numbers of computing equipment. Thus, one characteristic may be
more influential than another characteristic in determining a deal
similarity.
[0041] With continued reference to the deals clustered based on
account similarity, a ratio of deals may be calculated, for
example, using the 10 most similar deals as long as each deal has a
similarity measure above 0.8. The foregoing values are provided
only by way of example, and are not intended to limit the scope of
the invention in any way. This approach may also be employed with
respect to the clusters that include the quote characteristics.
That is, a ratio of deals may be determined based upon a deal
similarity that was determined with reference to the
characteristics of the accounts respectively associated with those
deals. When a new quote is generated, referring again to deals
clustered based on account similarity, the new quote may be
assigned to one of the deal clusters, and, as in the aforementioned
illustrative case of 10 most similar deals, a ratio of deals
calculated for that deal cluster to which the new quote was
assigned.
[0042] The quote characteristics, and/or account characteristics,
along with information concerning the engineered features, that is,
the features available in products offered by the sales entity, may
then be provided as inputs to the machine learning model 102. In
some embodiments, the machine learning model 102 may comprise, or
employ, a random forest learning method, which may also be referred
to as a random decision forest. The machine learning model 102 may
then process the aforementioned, and/or other, inputs to generate,
as an output, a probability that a new/adjusted proposal, or quote,
110 will be approved by the target customer entity.
[0043] B.2 Aspects of Example Methods for Quote Adjustment
[0044] With continued reference to the example of FIG. 1, details
are provided concerning an example framework for adjusting a quote
with the purpose of increasing the probability of acceptance. Such
adjustments may be made on various bases.
[0045] In some embodiments, a quote, or proposal, may be adjusted
based on product specs and hierarchy. For example, when the
probability of a quote to be approved has been determined by the
machine learning model 102 to be unacceptably low, the sales entity
may set some constraints and try to adjust the quote to generate a
higher probability.
[0046] For example, the sales entity may be able to define what
products he would like to keep intact, that is, products whose
configuration or characteristics will remain static, and which
products can be changed. The sales entity may also define, for
example, if one product should be replaced with a similar one in
order to give further discount, if the sales entity would like to
upgrade a certain product, or if the sales entity would like to
simply try and increase the probability of acceptance without
specifying whether or not the product should be
upgraded/replaced.
[0047] Using any constraints specified by the sales entity, or
otherwise applied, a search algorithm, which may be an element of
the machine learning model 102, may be employed in an attempt to
generate a revised quote with a relatively higher likelihood of
acceptance by the target customer. Any of various different
approaches may employed in the generation of a revised quote. One
approach would be for the search algorithm to identify all possible
options for modifying the initial quote. Suppose, for example, the
sales entity would like to upgrade a certain laptop, but only up to
a certain price. In this example, the framework may evaluate all
available possibilities to modify the laptop configuration without
exceeding that price. Such modification possibilities might
include, for example, ({better RAM}, {better CPU}, and {better RAM,
better CPU}). If each of these modifications, some of them, or all
of them, could be implemented without exceeding the price
constraint, the machine learning model 102 may determine the
probability of acceptance for each modification, and then select
the modification(s) that correspond to the highest probability of
acceptance.
[0048] As another example, the search algorithm may consider
various reconfiguration options in a serial manner. That is, the
search algorithm may evaluate one possible change to the quote per
iteration of a search, and check if the change that was made
generated a high probability of acceptance or not. Continuing with
the previous example, the option of including a better RAM may
first be evaluated. If additional costs of upgrading the RAM fall
within an acceptable range, the probability of acceptance will be
determined, but if the RAM upgrade cost is not acceptable, another
iteration may be performed to evaluate the advisability of, for
example, upgrading the CPU. Note that the various upgrade options
may all have the same relative weight, or different respective
weights. Thus, for example, an option that exceeds the price
constraint by a small amount, but is heavily weighted, may
nonetheless be considered for inclusion in the quote. The
weightings may be assigned by the customer entity and/or the sales
entity.
[0049] With reference now to FIG. 2, another illustrative example
is provided. In the example scenario 300 of FIG. 2, upgrading the
RAM would exceed the additional cost that was defined. On the other
hand, upgrading the CPU would not exceed the additional cost and
the probability of acceptance is high. Thus, in scenario 300, the
quote may be revised to include a CPU upgrade, but may not be
revised to include a RAM upgrade.
[0050] Turning next to FIG. 3, another example scenario 400 is
indicated. In this example, upgrading the RAM in the quote does not
exceed the additional costs. However, even with this upgrade, the
probability of acceptance is still low, and likewise for a CPU
upgrade. Since neither the RAM upgrade, nor the CPU upgrade,
considered alone has a positive effect on the probability of
acceptance, consideration may be given to determining how that
probability might be affected if both upgrades were included in the
quote. In this case, the new cost is still in the given range, and
the probability of acceptance is high. Thus, the sales entity may
update the quote to include both the RAM upgrade and the CPU
upgrade.
[0051] The foregoing example scenarios concerned possible
modifications to a quote based upon product specifications and an
associated hierarchy. Some embodiments may additionally, or
alternatively, provide for quote modifications based on historical
data, such as historical data relating to the behavior of customers
and/or prospective customers.
[0052] Collecting this kind of data, such as from e-commerce
websites, may be accomplished using clickstream analysis, for
example. From the collected data, samples may be extracted
indicating where a prospective customer examined a certain product
but eventually decided to buy a different one.
[0053] Another example is when there are two similar products in
the shopping cart of a prospective customer, but eventually one is
removed. Still another useful data source for this kind of behavior
is the pricing data set, where a quote with a certain configuration
is suggested to the customer, denied, and then followed by a
different suggestion with a different configuration which gets
approved by the customer.
[0054] After proposing alternative products to the product that is
currently being suggested to the customer, constraints imposed by
the sales entity may be introduced into the process to filter out
irrelevant products. Next, the similarity between the current
product and its alternatives may be examined. The similarity may be
calculated, or otherwise determined, based on, for example, the
specifications of the product, using cosine similarity and/or other
techniques. It is noted that cosine similarity, and/or comparable
techniques, may also be used to determine deal similarity, such as
was discussed in connection with FIG. 1. After the similarity
analysis has been performed, and similarity scores determined for
the various products, a selected group, such as 3 for example, of
the most similar products may be evaluated using the machine
learning model 102, and the product that is determined to have the
highest probability of acceptance may be proposed to the
customer.
[0055] B.3 Further Points
[0056] In view of the present disclosure, it will be apparent that
conventional approaches for adjusting a quote or proposal are
flawed for a variety of reasons. For example, such adjustments are
often based on little more than the intuition of a salesman as to
whether or such not the adjustment will be accepted by a
prospective consumer. As another example, a mis-assessment on the
part of the salesman may result in a lost opportunity, such as a
lost sale.
[0057] Finally, making adjustments to a proposal, while at the same
time attempting to maintain an acceptable profit margin for the
proposal, is a complex process that is beyond the ability of a
human to do without introducing the possibility of human error, and
also beyond the ability of a human to perform in a timely manner.
Such embodiments may, for example, generate an updated quote and/or
updated probability almost immediately, such as in about 5 seconds
or less and possibly as quickly as about 1 second or less, after
input relating to the updated quote and/or updated probability has
been received. Thus, embodiments of the invention do not merely
organize human behavior but, instead, such embodiments implement
functionality that a human is not capable of performing,
practically or in any way. Accordingly, such functionality is
performed by a computing system, as disclosed herein.
[0058] At least some embodiments may, but do not necessarily,
address such concerns by way of various functionalities. For
example, at least some embodiments may create new/modified
proposals on a systematic basis using factors other than, or in
addition to, human intuition. As well, some embodiments may employ
various types of historical data, and/or account characteristics,
product specifications and hierarchies (see, e.g., FIGS. 2 and 3),
as bases for creating a new/modified proposal with an acceptable
probability of acceptance by a user. Further, some embodiments may
provide for creation/modification of a proposal while maintaining a
particular margin, which the sales entity may not be aware of,
associated with the proposal. For example, the margin may be
imposed by an accounting or other department at an enterprise, but
not revealed to the sales personnel. As a final example, some
embodiments of the invention may elevate customer satisfaction by
quickly identifying a proposal with a high probability of
acceptance.
C. Example Methods
[0059] It is noted with respect to the example method of Figure(s)
XX that any of the disclosed processes, operations, methods, and/or
any portion of any of these, may be performed in response to, as a
result of, and/or, based upon, the performance of any preceding
process(es), methods, and/or, operations. Correspondingly,
performance of one or more processes, for example, may be a
predicate or trigger to subsequent performance of one or more
additional processes, operations, and/or methods. Thus, for
example, the various processes that may make up a method may be
linked together or otherwise associated with each other by way of
relations such as the examples just noted.
[0060] Directing attention now to FIG. 4, details are provided
concerning methods and processes for generating a new or modified
proposal, where one example of such a method is denoted generally
at 500. The example method 500 may be performed in whole, or in
party, by/at an enterprise, which may include a machine learning
model, associated with which one or more sales entities are
associated. In some cases, the example method 500 may be performed
cooperatively by one or more sales entities and an enterprise. In
some instances, the method 500 may be performed in whole or in part
by a quotation platform that comprises, or consists of, a machine
learning model. However, no particular implementation of the method
500 is required, and the foregoing are presented only by way of
example.
[0061] The example method 500 may begin with instantiation, such as
by a sales entity, of a quote process 502. The process 502 may
involve, for example instantiation at a mobile device, or
instantiation of a web app using a web browser. The sales entity or
other user may then provide various input which may be received 504
by a quotation software platform or other entity. The input
received 504 may comprise, for example, characteristics of one or
more prior quotations that are similar in one or more respects to
the quotation to be generated by the method 500, characteristics of
an account for whom the quotation is to be generated, historical
information concerning the prior quotations, particular users
associated with the account, and/or the account itself.
[0062] Next, a new or modified quotation may be generated 506, such
as by a sales entity for example, or automatically in some cases.
The new/modified quotation may identify, for example, an account
for whom the quotation was generated 506, the products/services
proposed to be provided to the account, and the price associated
with provision of those products/services. The quote may be
generated, in whole or in part, based on the input received at 504,
and may be configured so as to provide at least a specified profit
margin if accepted by a customer. As well, the generated quote 506,
whether new or modified, may serve as an input to a machine
learning model which may then determine 508 the probability that
the quote will be accepted by the customer. This determination 508
may be based, for example, on the input received at 504.
[0063] If the probability that the quote will be accepted is
evaluated 510 as unacceptable, the method 500 may return to 506,
and the quote modified in an attempt to improve the probability
that the quote will be accepted by the customer. On the other hand,
if the probability that the quote will be accepted is evaluate 510
as being acceptable, or acceptably high, the method 500 may advance
to 512 where the quote is presented to the customer.
[0064] Regardless of whether the quote that is generated at 506 is
determined 510 to have a sufficient probability of acceptance, that
quote may be stored as historical data that may be used as an input
504, for example, to generate 506 one or more future quotes and/or
as a basis for determining 508 the probability that a future quote
will be accepted by a customer. The quote that is ultimately
presented to the customer 512 may be employed in these same ways as
well.
D. Further Example Embodiments
[0065] Following are some further example embodiments of the
invention. These are presented only by way of example and are not
intended to limit the scope of the invention in any way.
[0066] Embodiment 1. A method, comprising: receiving a quote for
provision of goods and/or services, and the quote concerns a
particular account; receiving information concerning
characteristics of the account identified in the quote; receiving
information concerning characteristics of the goods and/or services
specified in the quote; generating a probability that the quote
will be approved by the account, and the probability is generated
based on the characteristics of the account and the characteristics
of the goods and/or services specified in the quote; when the
probability is below a threshold, generating an adjusted quote
based on input received; and generating an updated probability
based on the adjusted quote.
[0067] Embodiment 2. The method as recited in embodiment 1, wherein
part of the method is performed by a machine learning model.
[0068] Embodiment 3. The method as recited in any of embodiments
1-2, wherein the probability and/or updated probability are
generated based on one or both of: a first deal cluster comprising
a plurality of deals and identifying common quote characteristics
among respective quotes that are associated with the deals; and/or
a second deal cluster comprising a plurality of deals and
identifying common account characteristics among respective
accounts associated with the deals.
[0069] Embodiment 4. The method as recited in embodiment 3, further
comprising assigning the updated quote to one of the deal clusters,
and updating a ratio of approved deals associated with the deal
cluster to which the updated quote is assigned.
[0070] Embodiment 5. The method as recited in any of embodiments
1-4, further comprising: presenting, to a user, possible
modifications to goods and/or services specified in the quote;
receiving, from the user, the input, and the input indicating
selection of one or more of the modifications; generating the
adjusted quote based on the input received from the user; and
providing, to the user, the updated probability.
[0071] Embodiment 6. The method as recited in embodiment 5, wherein
the modifications are presented in serial form, or hierarchical
form.
[0072] Embodiment 7. The method as recited in any of embodiments
1-6, wherein the updated probability and/or the updated quote are
generated almost immediately after receipt of the input from the
user.
[0073] Embodiment 8. The method as recited in any of embodiments
1-7, further comprising: generating a first deal cluster comprising
a plurality of deals and identifying common quote characteristics
among respective quotes that are associated with the deals; and/or
generating a second deal cluster comprising a plurality of deals
and identifying common account characteristics among respective
accounts associated with the deals, and generation of the updated
probability is based on the characteristics identified in the first
deal cluster and/or the characteristics identified in the second
deal cluster.
[0074] Embodiment 9. The method as recited in any of embodiments
1-8, wherein a margin associated with the quote is maintained in
the updated quote.
[0075] Embodiment 10. The method as recited in any of embodiments
1-9, wherein the updated probability is higher than the
probability.
[0076] Embodiment 11. A method for performing any of the
operations, methods, or processes, or any portion of any of these,
disclosed herein.
[0077] Embodiment 12. A non-transitory storage medium having stored
therein instructions that are executable by one or more hardware
processors to perform operations comprising the operations of any
one or more of embodiments 1 through 11.
E. Example Computing Devices and Associated Media
[0078] The embodiments disclosed herein may include the use of a
special purpose or general-purpose computer including various
computer hardware or software modules, as discussed in greater
detail below. A computer may include a processor and computer
storage media carrying instructions that, when executed by the
processor and/or caused to be executed by the processor, perform
any one or more of the methods disclosed herein, or any part(s) of
any method disclosed.
[0079] As indicated above, embodiments within the scope of the
present invention also include computer storage media, which are
physical media for carrying or having computer-executable
instructions or data structures stored thereon. Such computer
storage media may be any available physical media that may be
accessed by a general purpose or special purpose computer.
[0080] By way of example, and not limitation, such computer storage
media may comprise hardware storage such as solid state disk/device
(SSD), RAM, ROM, EEPROM, CD-ROM, flash memory, phase-change memory
("PCM"), or other optical disk storage, magnetic disk storage or
other magnetic storage devices, or any other hardware storage
devices which may be used to store program code in the form of
computer-executable instructions or data structures, which may be
accessed and executed by a general-purpose or special-purpose
computer system to implement the disclosed functionality of the
invention. Combinations of the above should also be included within
the scope of computer storage media. Such media are also examples
of non-transitory storage media, and non-transitory storage media
also embraces cloud-based storage systems and structures, although
the scope of the invention is not limited to these examples of
non-transitory storage media.
[0081] Computer-executable instructions comprise, for example,
instructions and data which, when executed, cause a general purpose
computer, special purpose computer, or special purpose processing
device to perform a certain function or group of functions. As
such, some embodiments of the invention may be downloadable to one
or more systems or devices, for example, from a website, mesh
topology, or other source. As well, the scope of the invention
embraces any hardware system or device that comprises an instance
of an application that comprises the disclosed executable
instructions.
[0082] Although the subject matter has been described in language
specific to structural features and/or methodological acts, it is
to be understood that the subject matter defined in the appended
claims is not necessarily limited to the specific features or acts
described above. Rather, the specific features and acts disclosed
herein are disclosed as example forms of implementing the
claims.
[0083] As used herein, the term `module` or `component` may refer
to software objects or routines that execute on the computing
system. The different components, modules, engines, and services
described herein may be implemented as objects or processes that
execute on the computing system, for example, as separate threads.
While the system and methods described herein may be implemented in
software, implementations in hardware or a combination of software
and hardware are also possible and contemplated. In the present
disclosure, a `computing entity` may be any computing system as
previously defined herein, or any module or combination of modules
running on a computing system.
[0084] In at least some instances, a hardware processor is provided
that is operable to carry out executable instructions for
performing a method or process, such as the methods and processes
disclosed herein. The hardware processor may or may not comprise an
element of other hardware, such as the computing devices and
systems disclosed herein.
[0085] In terms of computing environments, embodiments of the
invention may be performed in client-server environments, whether
network or local environments, or in any other suitable
environment. Suitable operating environments for at least some
embodiments of the invention include cloud computing environments
where one or more of a client, server, or other machine may reside
and operate in a cloud environment.
[0086] With reference briefly now to FIG. 5, any one or more of the
entities disclosed, or implied, by FIGS. 1-4 and/or elsewhere
herein, may take the form of, or include, or be implemented on, or
hosted by, a physical computing device, one example of which is
denoted at 600. As well, where any of the aforementioned elements
comprise or consist of a virtual machine (VM), that VM may
constitute a virtualization of any combination of the physical
components disclosed in FIG. 5.
[0087] In the example of FIG. 5, the physical computing device 600
includes a memory 602 which may include one, some, or all, of
random access memory (RAM), non-volatile memory (NVM) 604 such as
NVRAM for example, read-only memory (ROM), and persistent memory,
one or more hardware processors 606, non-transitory storage media
608, UI device 610, and data storage 612. One or more of the memory
components 602 of the physical computing device 600 may take the
form of solid state device (SSD) storage. As well, one or more
applications 614 may be provided that comprise instructions
executable by one or more hardware processors 606 to perform any of
the operations, or portions thereof, disclosed herein.
[0088] Such executable instructions may take various forms
including, for example, instructions executable to perform any
method or portion thereof disclosed herein, and/or executable by/at
any of a storage site, whether on-premises at an enterprise, or a
cloud computing site, client, datacenter, data protection site
including a cloud storage site, or backup server, to perform any of
the functions disclosed herein. As well, such instructions may be
executable to perform any of the other operations and methods, and
any portions thereof, disclosed herein.
[0089] The present invention may be embodied in other specific
forms without departing from its spirit or essential
characteristics. The described embodiments are to be considered in
all respects only as illustrative and not restrictive. The scope of
the invention is, therefore, indicated by the appended claims
rather than by the foregoing description. All changes which come
within the meaning and range of equivalency of the claims are to be
embraced within their scope.
* * * * *
References